Human activity detection and recognition is a challenging task.Video surveillance can benefit greatly by advances in Internet of Things(IoT)and cloud computing.Artificial intelligence IoT(AIoT)based devices form the b...Human activity detection and recognition is a challenging task.Video surveillance can benefit greatly by advances in Internet of Things(IoT)and cloud computing.Artificial intelligence IoT(AIoT)based devices form the basis of a smart city.The research presents Intelligent dynamic gesture recognition(IDGR)using a Convolutional neural network(CNN)empowered by edit distance for video recognition.The proposed system has been evaluated using AIoT enabled devices for static and dynamic gestures of Pakistani sign language(PSL).However,the proposed methodology can work efficiently for any type of video.The proposed research concludes that deep learning and convolutional neural networks give a most appropriate solution retaining discriminative and dynamic information of the input action.The research proposes recognition of dynamic gestures using image recognition of the keyframes based on CNN extracted from the human activity.Edit distance is used to find out the label of the word to which those sets of frames belong to.The simulation results have shown that at 400 videos per human action,100 epochs,234×234 image size,the accuracy of the system is 90.79%,which is a reasonable accuracy for a relatively small dataset as compared to the previously published techniques.展开更多
Statistical machine translation for low-resource language suffers from the lack of abundant training corpora. Several methods, such as the use of a pivot language, have been proposed as a bridge to translate from one ...Statistical machine translation for low-resource language suffers from the lack of abundant training corpora. Several methods, such as the use of a pivot language, have been proposed as a bridge to translate from one language to another. However, errors will accumulate during the extensive translation pipelines. In this paper, we propose an approach to low-resource language translation by exploiting the pronunciation correlations between languages. We find that the pronunciation features can improve both Chinese-Vietnamese and Vietnamese- Chinese translation qualities. Experimental results show that our proposed model yields effective improvements, and the translation performance (bilingual evaluation understudy score) is improved by a maximum value of 1.03.展开更多
Owing to advanced storage and communication capabilities today, smart devices have become the basic interface between individuals and their surrounding environment. In particular, massive devices connect to one other ...Owing to advanced storage and communication capabilities today, smart devices have become the basic interface between individuals and their surrounding environment. In particular, massive devices connect to one other directly in a proximity area, thereby enabling abundant Proximity Services(Pro Se), which can be classified into two categories: public safety communication and social discovery. However, two challenges impede the quick development and deployment of Pro Se applications. From the viewpoint of networking, no multi-hop connectivity functionality component can be directly operated on commercially off-the-shelf devices, and from the programming viewpoint, an easily reusable development framework is lacking for developers with minimal knowledge of the underlying communication technologies and connectivity. Considering these two issues, this paper makes a twofold contribution. First, a multi-hop mesh networking based on Bluetooth Low Energy(BLE) is implemented,in which a proactive routing mechanism with link-quality(i.e., received signal strength indication) assistance is designed. Second, a Pro Se development framework called BLE Mesh is designed and implemented, which can provide significant benefits for application developers, framework maintenance professionals, and end users. Rich application programming interfaces can help developers to build Pro Se apps easily and quickly. Dependency inversion principle and template method pattern allow modules in BLE Mesh to be loosely coupled and easy to maintain and update. Callback mechanism enables modules to work smoothly together and automation processes such as registration, node discovery, and messaging are employed to offer nearly zero-configuration for end users.Finally, based on the designed Pro Se development kit, a public safety communications app called Quote Send App is built to distribute emergency information in close area without Internet access. The process illustrates the easy usability of BLE Mesh to develop Pro Se apps.展开更多
This study, based on the theory of equivalence relations, proposes a novel multilevel index model for decentralized service repositories to eliminate redundant information and enhance the time-management quality of th...This study, based on the theory of equivalence relations, proposes a novel multilevel index model for decentralized service repositories to eliminate redundant information and enhance the time-management quality of the service retrieval process of the service repository architecture. An efficient resource discovery algorithm based on Discrete Hash Tables is presented to enable efficient and effective retrieval services among different distributed repositories. The performance of the proposed model and the supporting algorithms have been evaluated in a distributed environment. Experimental results validate the effectiveness of our proposed indexing model and search algorithm.展开更多
文摘Human activity detection and recognition is a challenging task.Video surveillance can benefit greatly by advances in Internet of Things(IoT)and cloud computing.Artificial intelligence IoT(AIoT)based devices form the basis of a smart city.The research presents Intelligent dynamic gesture recognition(IDGR)using a Convolutional neural network(CNN)empowered by edit distance for video recognition.The proposed system has been evaluated using AIoT enabled devices for static and dynamic gestures of Pakistani sign language(PSL).However,the proposed methodology can work efficiently for any type of video.The proposed research concludes that deep learning and convolutional neural networks give a most appropriate solution retaining discriminative and dynamic information of the input action.The research proposes recognition of dynamic gestures using image recognition of the keyframes based on CNN extracted from the human activity.Edit distance is used to find out the label of the word to which those sets of frames belong to.The simulation results have shown that at 400 videos per human action,100 epochs,234×234 image size,the accuracy of the system is 90.79%,which is a reasonable accuracy for a relatively small dataset as compared to the previously published techniques.
基金This work was supported by Beijing Science and Technology Program (No. D141100000514001), National Natural Science Foundation of China (No. 51372133), and National Program on Key Basic Research Project (Nos. 2011CB013000 and 2014CB932401)
基金supported by the National key Basic Research and Development(973)Program of China(No.2013CB329303)the National Natural Science Foundation of China(Nos.61502035,61132009,and 61671064)Beijing Advanced Innovation Center for Imaging Technology(No.BAICIT-2016007)
文摘Statistical machine translation for low-resource language suffers from the lack of abundant training corpora. Several methods, such as the use of a pivot language, have been proposed as a bridge to translate from one language to another. However, errors will accumulate during the extensive translation pipelines. In this paper, we propose an approach to low-resource language translation by exploiting the pronunciation correlations between languages. We find that the pronunciation features can improve both Chinese-Vietnamese and Vietnamese- Chinese translation qualities. Experimental results show that our proposed model yields effective improvements, and the translation performance (bilingual evaluation understudy score) is improved by a maximum value of 1.03.
基金supported by the National Natural Science Foundation of China(No.61171092)Jiangsu Educational Bureau Project(No.14KJA510004)NUPTSFs(Nos.NY215177 and NY217089)
文摘Owing to advanced storage and communication capabilities today, smart devices have become the basic interface between individuals and their surrounding environment. In particular, massive devices connect to one other directly in a proximity area, thereby enabling abundant Proximity Services(Pro Se), which can be classified into two categories: public safety communication and social discovery. However, two challenges impede the quick development and deployment of Pro Se applications. From the viewpoint of networking, no multi-hop connectivity functionality component can be directly operated on commercially off-the-shelf devices, and from the programming viewpoint, an easily reusable development framework is lacking for developers with minimal knowledge of the underlying communication technologies and connectivity. Considering these two issues, this paper makes a twofold contribution. First, a multi-hop mesh networking based on Bluetooth Low Energy(BLE) is implemented,in which a proactive routing mechanism with link-quality(i.e., received signal strength indication) assistance is designed. Second, a Pro Se development framework called BLE Mesh is designed and implemented, which can provide significant benefits for application developers, framework maintenance professionals, and end users. Rich application programming interfaces can help developers to build Pro Se apps easily and quickly. Dependency inversion principle and template method pattern allow modules in BLE Mesh to be loosely coupled and easy to maintain and update. Callback mechanism enables modules to work smoothly together and automation processes such as registration, node discovery, and messaging are employed to offer nearly zero-configuration for end users.Finally, based on the designed Pro Se development kit, a public safety communications app called Quote Send App is built to distribute emergency information in close area without Internet access. The process illustrates the easy usability of BLE Mesh to develop Pro Se apps.
基金supported by the National Natural Science Foundation of China(Nos.61502209 and 61502207)Postdoc Funds of China and Jiangsu Province(Nos.2015M580396 and 1501023A)the Jiangsu University Foundation(No.5503000049)
文摘This study, based on the theory of equivalence relations, proposes a novel multilevel index model for decentralized service repositories to eliminate redundant information and enhance the time-management quality of the service retrieval process of the service repository architecture. An efficient resource discovery algorithm based on Discrete Hash Tables is presented to enable efficient and effective retrieval services among different distributed repositories. The performance of the proposed model and the supporting algorithms have been evaluated in a distributed environment. Experimental results validate the effectiveness of our proposed indexing model and search algorithm.